The storage system is one of the most limiting aspects of performance of modern enterprise computing systems. Performance of hard drive based storage is determined by seek time and time for half rotation. The performance is increased by decreasing seek time and decreasing rotational latency. However, there are limits on how fast a drive may spin. The fastest contemporary drives are reaching 15,000 rpm.
FIG. 1 illustrates a system 100 in accordance with the prior art. In the system 100, at least one computer 102-108 is coupled to a host controller 110 and 112. The host controllers 110 and 112 are coupled to a plurality of disks 114-120.
Often, the system 100 is configured as redundant array of independent disks (RAID)-1, storing mirrored content of the disks 114-116 in the disks 118-120. The disks 114-116 are said to be mirrored by the disks 118-120.
Increased reliability of the computer system is achieved by duplicating the disks 114-116, the host controllers 110 and connections therebetween. Therefore, a reliable computer system is able operate at least in presence of single failure of the disks 114-120, the RAID controllers 110 and 112, the computers 102-108, and the connections therebetween. However, storage system performance may still be inadequate using the system 100. Additionally, increasing the performance of such system is currently costly and often times is not feasible.
Furthermore, one limiting aspect of current storage systems is the fact that many types of storage devices exhibit a limited lifetime. For example, a lifetime of non-volatile memory such as flash is reduced each time it is erased and re-written. Over time and thousands of erasures and re-writes, such storage systems may become less and less reliable.
There is thus a need for addressing these and/or other issues associated with the prior art.